Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A robotic device, comprising: a chassis including a set of wheels; a motor to drive the set of wheels; a battery to power the robotic device; a controller in communication with the motor and wheels, the controller including a processor operable to control the motor and wheels to steer movement of the robotic device; a camera; and, a tangible, non-transitory, machine readable medium storing instructions that when executed by an image processor effectuates operations comprising: causing the camera to capture one or more images of an environment of the robotic device; receiving, with the image processor, one or more multidimensional arrays including at least one parameter that describes a feature included in the one or more images, wherein values of the at least one parameter correspond with pixels of a corresponding one or more images of the feature; determining, with the image processor, an amount of asymmetry of the feature in the one or more images based on at least a portion of the values of the at least one parameter; and, transmitting, with the image processor, a signal to the processor of the controller to adjust a heading of the robotic device by an amount proportional to the amount of asymmetry of the feature, and wherein the robotic device maintains its heading while moving along a movement path within the environment until it either reaches a border of the environment or the image processor detects asymmetry of the feature in the one or more images.
2. The robotic device of claim 1 , wherein determining the amount of asymmetry comprises comparing at least a portion of the values of the at least one parameter corresponding with pixels of a first portion of an image of the one or more images with at least a portion of the values of the at least one parameter corresponding with pixels of a second portion of the image.
3. The robotic device of claim 2 , wherein the first portion of the image and the second portion of the image correspond to a division of the image about a vertical line or horizontal line.
4. The robotic device of claim 1 , wherein the movement path is perpendicular or parallel to a surface of the feature.
5. The robotic device of claim 1 , wherein the feature is projected onto a surface perpendicular or parallel to the movement path of the robotic device using one or more light emitters.
6. The robotic device of claim 1 , wherein determining the amount of asymmetry comprises: counting a number of columns or rows of pixels found between a division line of an image of the one or more images and a first column or row of pixels containing at least one pixel with brightness intensity above a predetermined threshold in a first and second direction from the division line, and subtracting a number of pixels from the other, the second direction being opposite the first, wherein a direction of the heading adjustment is indicated by a sign of the amount of asymmetry, wherein a positive and a negative amount of asymmetry indicate opposite directions.
7. The robotic device of claim 6 , wherein determining the heading adjustment comprises multiplying the amount of asymmetry by a predetermined ratio of heading adjustment per pixel.
8. The robotic device of claim 1 , wherein determining the amount of asymmetry comprises comparing the overlap of pixels between two images captured consecutively.
9. The robotic device of claim 1 , wherein determining the amount of asymmetry comprises: identifying a position of a first point of the projected feature in an image of the one or more images by a first set of coordinates and a position of a second point of the projected feature in the image by a second set of coordinates, the second point symmetrically corresponding with the first point when the heading is accurate, and, determining the distance of the first point and second point from a division line in the image and subtracting one distance from the other.
10. The robotic device of claim 9 , wherein determining the heading adjustment comprises multiplying the amount of asymmetry by a predetermined ratio of heading adjustment per unit of distance.
11. The robotic device of claim 1 , wherein the feature comprises a line, a curve, a polygon, one or more points, an edge, or a corner.
12. The robotic device of claim 1 , wherein the camera continuously captures images of the feature and the image processor continuously monitors and determines any adjustment required in the heading of the robotic device based on the captured images.
13. A method for guiding a robotic device, comprising: capturing, with a camera, one or more images of a feature of an environment; receiving, with an image processor, the one or more images of the feature; determining, with the image processor, an amount of asymmetry of the feature in the one or more images based on at least a portion of the pixels in the one or more images; and, transmitting, with the image processor, a signal to a controller in communication with a wheel motor of a robotic device to adjust the heading of the robotic device by an amount proportional to the amount of asymmetry of the feature, wherein the controller maintains the heading of the robotic device while moving along a movement path until the robotic device either reaches a border or the image processor detects asymmetry of the feature in the one or more images.
14. The method of claim 13 , wherein the feature is projected onto a surface using one or more light emitters.
15. The method of claim 13 , wherein the camera continuously captures images of the environment and the image processor constantly monitors and determines any adjustment required in the heading of the robotic device based on the captured images.
16. The method of claim 13 , wherein the image processor uses at least a portion of the one or more images to determine a distance to a surface of the feature.
17. The method of claim 13 , wherein adjusting the heading of the robotic device comprises: counting, with the image processor, a number of columns or rows of pixels between a division line of an image of the one or more images and a first column or row of pixels containing at least one pixel with brightness intensity above a predetermined threshold in a first and second direction from the division line and subtracting one number of pixels from the other to obtain the amount of asymmetry, the second direction being opposite from the first; and, determining, with the image processor, the heading adjustment by multiplying the amount of asymmetry by a predetermined ratio of heading adjustment per pixel.
This invention relates to autonomous navigation of robotic devices using visual asymmetry detection. The problem addressed is ensuring precise heading adjustments for robotic devices, such as drones or ground vehicles, to maintain accurate navigation without relying solely on external positioning systems. The solution involves analyzing images captured by the robotic device to detect visual asymmetry, which is then used to adjust the device's heading. The method processes one or more images to identify a division line, such as a horizon or a reference marker, and then evaluates pixel brightness intensity in two opposite directions from this line. The system counts the number of columns or rows of pixels between the division line and the first column or row containing pixels with brightness above a predetermined threshold in both directions. The difference between these counts determines the amount of asymmetry. The heading adjustment is calculated by multiplying this asymmetry by a predetermined ratio of heading adjustment per pixel, ensuring the robotic device corrects its path based on visual feedback. This approach enables real-time, self-correcting navigation by leveraging image processing to detect and compensate for deviations from the intended heading.
18. The method of claim 13 , wherein adjusting the heading of the robotic device comprises: identifying, with the image processor, a position of a first point of the feature in an image of the one or more images by a first set of coordinates and a position of a second point of the feature in the image by a second set of coordinates, the second point symmetrically corresponding with the first point when the heading is accurate; determining, with the image processor, the distance of the first point and the second point from a division line of the image and subtracting one distance from the other to obtain the amount of asymmetry; and, determining, with the image processor, the heading adjustment by multiplying the amount of asymmetry by a predetermined ratio of heading adjustment per unit of distance.
19. The method of claim 13 , wherein the feature comprises one or more of a line, a curve, a polygon, one or more points, an edge, a corner, a wall, and a floor.
20. The method of claim 13 , wherein determining the amount of asymmetry of the feature comprises comparing an overlap of pixels between two images captured consecutively.
This invention relates to image processing techniques for analyzing asymmetry in features captured in consecutive images. The problem addressed is the need for accurate and efficient detection of asymmetrical features in sequential image data, which is critical in applications such as medical imaging, quality control, and surveillance. The method involves capturing two images in quick succession and analyzing the overlap of pixels between them to determine the degree of asymmetry. The process begins by obtaining a first image of a feature, followed by a second image of the same feature under similar conditions. The system then aligns the two images to account for any positional or rotational differences. Next, the method compares the pixel data of the two images to identify regions where pixels overlap and where they do not. The overlap is quantified to assess the asymmetry of the feature. A higher degree of overlap indicates symmetry, while significant differences suggest asymmetry. The method may also include adjusting the alignment parameters iteratively to refine the asymmetry measurement. This approach ensures precise detection of asymmetrical features, which is useful in applications requiring high accuracy, such as medical diagnostics or industrial inspections. The technique is particularly effective in scenarios where subtle changes in feature symmetry need to be detected reliably.
Unknown
January 26, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.